These are chat archives for ethersphere/orange-lounge

26th
Mar 2018
lash
@nolash
Mar 26 07:44
@zelig after standup? (it's daily now?)
Viktor Trón
@zelig
Mar 26 07:45
ok @nolash indeed it is daily now
Balint Gabor
@gbalint
Mar 26 07:45
@zelig @nolash it is good for me too
Viktor Trón
@zelig
Mar 26 07:45
and is at 4pm CET except for wednesday when it is 3pm
lash
@nolash
Mar 26 07:46
then the calendar is correct! Good.
lash
@nolash
Mar 26 07:52
@gbalint @zelig mentioned you working on config refactor? Where is the PR on it? I didn't notice
Balint Gabor
@gbalint
Mar 26 07:55
config refactor? I don’t know what you mean...
i’m working on chunker refactor and encryption
Balint Gabor
@gbalint
Mar 26 07:59
Oh, I just changed ChunkerParams in config, I think there is no conflict here
I’m pushing it today, will send it to you
lash
@nolash
Mar 26 10:46
@janos re https://github.com/ethersphere/go-ethereum/pull/337#pullrequestreview-106819024 I agree it's may seem a bit messy. My objective was to collect identical parameters that apply to all stores in one base structure. I did not endeavour to create a good permanent solutions since, as you mentioned, I was under the impression that the store organization was about to be changed anyway.
Balint Gabor
@gbalint
Mar 26 10:47
@nolash my encryption PR (with the config changes) ethersphere/go-ethereum#327
Janoš Guljaš
@janos
Mar 26 10:48
Hi, I understand completely.
lash
@nolash
Mar 26 10:48
However, my intutition tells me that it makes sense to have some notion of a base store object across the stores.
Janoš Guljaš
@janos
Mar 26 10:48
Just wanted to mention that it is a sign that storage package needs a bit of cleaning up.
lash
@nolash
Mar 26 10:49
And to my thinking it follows that the initialization of these parameters should be predictable and recognizable across stores too.
@gbalint ok I will review this
Janoš Guljaš
@janos
Mar 26 10:52
Yes, I would go further and reorganize the stores so that we do not need to do params this way. And this change is planed to be done.
lash
@nolash
Mar 26 10:52
@janos cool
lash
@nolash
Mar 26 17:07
@nagydani could you please postthe link of the latest encryption spec; the one the swarm-network-rewrite-encryption implementation is built on? Thanks
Daniel A. Nagy
@nagydani
Mar 26 17:52
Sure.
lash
@nolash
Mar 26 17:54
Cheers
Daniel A. Nagy
@nagydani
Mar 26 17:54
Actually, this is not entirely accurate either.
The SHA3 CTR is not fully specified.
The way it actually works is SHA3(SHA3(key|counter)), where | is the concatenation operator. key is the 256 bit encryption key, counter is a 32-bit counter, which, for the first 8 bytes is -1 (i.e. 0xFFFFFFFF), and is increased by one for each subsequent 32-byte slices .
Actually, thanks for reminding me that I should update the specification to what we actually implemented! Also, it should be part of the package documentation.
lash
@nolash
Mar 26 18:38
Cool